Re: [PATCH v7u1 10/31] x86, 64bit: Don't set max_pfn_mapped wrongvalue early on native path

[Stefano Stabellini]

On Tue, 15 Jan 2013, Konrad Rzeszutek Wilk wrote:
> On Tue, Jan 15, 2013 at 01:48:45PM +0000, Stefano Stabellini wrote:
> > On Fri, 4 Jan 2013, Yinghai Lu wrote:
> > > We are not having max_pfn_mapped set correctly until init_memory_mapping.
> > > 
> > > so don't print it initial value for 64bit
> > > 
> > > Also need to use KERNEL_IMAGE_SIZE directly for highmap cleanup.
> > > 
> > > Signed-off-by: Yinghai Lu <yinghai@xxxxxxxxxx>
> > > ---
> > >  arch/x86/kernel/head64.c |    3 ---
> > >  arch/x86/kernel/setup.c  |    2 ++
> > >  arch/x86/mm/init_64.c    |    6 +++++-
> > >  3 files changed, 7 insertions(+), 4 deletions(-)
> > > 
> > > diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
> > > index a3fc233..7061d8b 100644
> > > --- a/arch/x86/kernel/head64.c
> > > +++ b/arch/x86/kernel/head64.c
> > > @@ -146,9 +146,6 @@ void __init x86_64_start_kernel(char * real_mode_data)
> > >  	/* clear bss before set_intr_gate with early_idt_handler */
> > >  	clear_bss();
> > >  
> > > -	/* XXX - this is wrong... we need to build page tables from scratch */
> > > -	max_pfn_mapped = KERNEL_IMAGE_SIZE >> PAGE_SHIFT;
> > > -
> > >  	for (i = 0; i < NUM_EXCEPTION_VECTORS; i++) {
> > >  #ifdef CONFIG_EARLY_PRINTK
> > >  		set_intr_gate(i, &early_idt_handlers[i]);
> > > diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
> > > index 63160c6..04797e78 100644
> > > --- a/arch/x86/kernel/setup.c
> > > +++ b/arch/x86/kernel/setup.c
> > > @@ -910,8 +910,10 @@ void __init setup_arch(char **cmdline_p)
> > >  	setup_bios_corruption_check();
> > >  #endif
> > >  
> > > +#ifdef CONFIG_X86_32
> > >  	printk(KERN_DEBUG "initial memory mapped: [mem 0x00000000-%#010lx]\n",
> > >  			(max_pfn_mapped<<PAGE_SHIFT) - 1);
> > > +#endif
> > >  
> > >  	reserve_real_mode();
> > >  
> > > diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
> > > index 9c5f2b1..98385a2 100644
> > > --- a/arch/x86/mm/init_64.c
> > > +++ b/arch/x86/mm/init_64.c
> > > @@ -394,10 +394,14 @@ void __init init_extra_mapping_uc(unsigned long phys, unsigned long size)
> > >  void __init cleanup_highmap(void)
> > >  {
> > >  	unsigned long vaddr = __START_KERNEL_map;
> > > -	unsigned long vaddr_end = __START_KERNEL_map + (max_pfn_mapped << PAGE_SHIFT);
> > > +	unsigned long vaddr_end = __START_KERNEL_map + KERNEL_IMAGE_SIZE;
> > >  	unsigned long end = roundup((unsigned long)_brk_end, PMD_SIZE) - 1;
> > >  	pmd_t *pmd = level2_kernel_pgt;
> > >  
> > > +	/* Xen has its own end somehow with abused max_pfn_mapped */
> > > +	if (max_pfn_mapped)
> > > +		vaddr_end = __START_KERNEL_map + (max_pfn_mapped << PAGE_SHIFT);
> > 
> > If you are going to put a comment like that in the code, could you
> > please at least add some useful details, rather than a generic
> > "somehow"? It doesn't seem very helpful to me or to any other hackers
> > looking at the code.
> 
> Hm I think I actually pointed out to him in the previous reviews how
> we alter it and that he should ingest some of those comments in this
> patch.

OK.
Still I think that altering max_pfn_mapped is the right thing to do for
Xen, because after all the last pfn mapped is different.
And "somehow" can't the best way to explain the reason for a change.



> > The issue is even described as a comment in the code at the beginning of
> > arch/x86/xen/mmu.c:xen_setup_kernel_pagetable:
> > 
> > /* max_pfn_mapped is the last pfn mapped in the initial memory
> >  * mappings. Considering that on Xen after the kernel mappings we
> >  * have the mappings of some pages that don't exist in pfn space, we
> >  * set max_pfn_mapped to the last real pfn mapped. */
> > 
> > Now if max_pfn_mapped is supposed to represent the last pfn mapped in
> > the initial memory mapping, then I think that the way Xen uses
> > max_pfn_mapped is actually correct.
> > 
> > 
> > The question is: has max_pfn_mapped actually changed meaning?
> > Because if it hasn't I don't see why you need this change.
> > 
> > 
> > 
> > >  	for (; vaddr + PMD_SIZE - 1 < vaddr_end; pmd++, vaddr += PMD_SIZE) {
> > >  		if (pmd_none(*pmd))
> > >  			continue;
> > > -- 
> > > 1.7.10.4
> > > 
> > > --
> > > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> > > Please read the FAQ at  http://www.tux.org/lkml/
> > > 
> 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/